worklog entry

Community Forums/Showcase/worklog entry

MadJack(Posted 2004) [#1]
What's up with the worklog titles?

And a new worklog for Tank Universal, if anyone's interested, by the way.


GNS(Posted 2004) [#2]
I've been wondering this as well. I believe I read something about a change to the worklogs system but I'm not sure.

As for the Tank Univeral worklog - great! I try and keep up with these as I find them interesting. :)


MadJack(Posted 2004) [#3]
GNS

Thanks for the interest.

I've gotta admit, I'm suffering from fatigue over this project.

Part of this has been from starting out with an incomplete design and allowing for feature creep in a big way. Obviously Blitz' ease of use is to blame for this ;-)

But I've put way too much time into it not to finish now.


GNS(Posted 2004) [#4]
I can certainly feel your pain. Nearly all of my projects end with me losing interest and generally wanting to move on to something else. I suppose for me the 'fun' lies in experimentation so I rarely keep interest long enough to mock up more than a few tests before moving on to something else.

For what it's worth I believe you've got something special on your hands with TU. I can honestly say I've played the demo atleast once everyday. One would think playing the same map over and over again would become boring (especially for someone like me who has a small attention span as it is!) but for some reason I just keep coming back for more. :)


MadJack(Posted 2004) [#5]
GNS

'I've played the demo at least once everyday'

You're kidding!

Actually I fire it up now and then as well just to remind myself where I'm going. I chose a tank game 'cause I think there's a (independant games) niche for a really well done, action packed tank game.

Also I noticed another bug in it the other day - artillery shells 'sticking' on laser turrets and not exploding.

However, I've decided to expand the fortress idea considerably and I'd like your thoughts about the following ideas;

On starting a level, there'll be no friendly units waiting for activation. The player firstly has to collect a number of 'projection gems' and take them to the fort to activate/create a hub.

A fortress commander is then airlifted in. He runs to a 'control room' in the fort, and calls in two harvesters. The harvesters then begin to collect gems (RTS style obviously). The commander uses collected gems to airlift in infantry, tanks and gun emplacements to defend the fort.

The player can help but this'll be autonomous. Meanwhile, the enemy's doing the same. The player can get out and harass the enemy, look for bonuses, defend etc, so there'll be plenty to do.

The goal is to reach a critical number of collected gems. Once done, then a massive incoming airdrop of friendly units occurs and they'll trundle off to attack the enemy fort (or vice-versa if the enemy have amassed enough gems first).

To sucessfully attack a fort, you'll need to scale the walls (FPS mode), fight to get to the enemy commander, terminate him and open the fortress gates allowing the tanks in to steal the hub key.

Get the hub key back to your fort and you've won.

Hurrah!


GNS(Posted 2004) [#6]
Haha, I wasn't exaggerating about me playing atleast once a day. It's hard to say exactly what pulls me back in. I'd say a combination of the atmosphere (nothing quite like seeing a glowing red mortar shell descend slowly onto a massive landscape!) and the general feeling of being apart of something 'bigger'. I've played far too many games where I felt as if I was controlling everything and my AI friends were little more than cannon fodder.

I noticed the 'mortar shell sticks to stuff' bug as well. Never mentioned it because it was kind of interesting getting to see the mortar shells up close. ;)

I agree that there seems to be a lack of well done tank games (especially in the indie world). For TU there's the added bonus of relatively few 'Tron' style games so the red and blue/digital world scheme is still unique.

As for the ideas, very interesting! Every idea mentioned would add to the atmosphere of an ongoing war in which the player isn't the only deciding factor (as I mentioned above, I quite like this). Having a commander who essentially controls the flow of the game would certainly give the game more depth. The player would actually have a reason (other than to aid his/her AI teammates) for fighting the opposing 'team'. I'm also fond of the use of the FPS view. Having to scale walls, besides being just plain cool, would give forts a reason for being around (no reason to have walls if the entrance is wide open!).


GNS(Posted 2004) [#7]
Doh! This died rather fast.


MadJack(Posted 2004) [#8]
Sorry GNS - haven't checked this thread for a few days - thanks for the feedback.

I want to give the idea that the player is on the run from powerful forces. So as a break, some levels will be a straight 'get from one end to the other end alive' deal.

So as some more examples of 'feature creep';

Level involves large vaulted rooms with ranks of regular columns (think the underside of a CPU and its pins), connected by narrow bridges across chasms. Bridges guarded by infantry. Enemy tanks, moving obstacles and squares that move tanks in alternative directions. Fog. Kind of your standard 2d tank game in 3d.

By the way, I think I'll avoid complex mazes. Deadends, identical passageways, having to doubleback, can't move on until it's solved?- no better way to suck the fun out of a game in my opinion. Bad game designer - no twinkie for you.

Or a simple maze dominated by high slowly rotating towers. If you're not under cover when a tower sweeps your postion, it sends out a powerful pulse starting from its base and moving quickly outwards to your position. Boom. Get to a tower's base unseen to deactivate it and move on.

Or a giant eyeball in a huge socket sunk into a pit in the landscape. Very big eyeball looking at player. The player must dodge across a landscape using sparse cover to get to the pit. Eyeball loosing death beams at player. Go down into pit, eyeball can't fire at this lower angle but fight enemy tanks released from socket. Take out struts holding eyeball in socket. Take out all struts and the eyeball floats up and explodes. Level over.

And of course I have to change the control scheme so that the RMB doesn't control the elevation of the gun - maybe that could be a shield recharge


GNS(Posted 2004) [#9]
I think you could blend the open spaces and maze-like city levels should the storyline provide for it. I do agree that the overused 'Hit button X to open door A which allows you to move into maze hall 1' would be annoying and not very fun. Going back to the maze-like city idea, it would be interesting to move from a level like the one featured in the demo (wide open spaces with tanks and air vehicles fighting in a "no man's land") to the innerworkings of a stratified city. It may even provide some extended use of the FPS mode (i.e. having to sabotage an enemy objective, on foot, by hiding from patrols in the alleyways and crevices that makeup such a city). But all of this is really me putting my personal spin on TU. :)

As for the RMB control scheme, I really can't be a good judge of this. I've gotten so used to using RMB to change the turret height that anything other than that would seem strange to me. In FPS mode it would make sense to change it to something more useful (whether it be secondary fire for a gun or use/access/interact with an object).


MadJack(Posted 2004) [#10]
GNS

'blend the open spaces and maze-like city levels'

Yep. Unfortunately though, it tends to throw up inconsistencies with the AI in that rules that work well for wide open spaces don't work as well for close quarters maze manouevering where the player is more likely to notice AI weaknesses. As the tanks make use of an underlying 128x128 array for A* pathfinding, I could probably institute some sort of 'tweaking' code that checks how 'encumbered' a tank is (by walls) and adjusts scan/engage/firing range for the unit. And although the tanks at present will try to hold a line and remain in contact with one another and call in reinforcements, they don't have any other manouevers. But then with a lot of units, extensive AI can become expensive. Another criticism of the TU demo is that enemy tanks don't react fast enough to shots and collisions.

The FPS mode is still rough, but getting there. Problem here is to avoid stranding the player in the badlands with no free tank nearby and nothing to do but a long trek home. To avoid this I'm considering dotting the landscape with hidden teleport booths.

'i.e. having to sabotage an enemy objective, on foot'

Yep. In conjunction with the fortress idea, I'm planning that while the player's fortress is gathering gems, the commander might give the player missions to do - which the player can complete or not. An example would be clearing out a hillside communications bunker - giving the player a full map. This would have a follow on in that the enemy fortress would receive occasional convoys of extra gems which would now become visible on the map and could be intercepted.

One last idea - a penalty for destroying friendly/civilian units - 'reputation'. Destroy enemy units and you gain reputation, destroy friendly units and you lose reputation. If it drops too far, you'll be a target for everyone -you'll need to destroy enough enemy targets then to regain status. Correspondingly, I want to introduce 'nukes' which are very rare but can clean out a fortress with one shot. But with reputation in effect, you wouldn't want to loose one off carelessly.


GNS(Posted 2004) [#11]
I imagined there would be some technical troubles with blending map types together. No matter though as each map type could essentially be a new 'level' (i.e. player wins demo-ish level with wide open spaces -> next level begins and places the player in a maze-city).

The idea of hidden teleport booths is an interesting solution. It may cause players to spend more time searching for the booths than actually treking home to find a new tank, though. There's also the problem of having players who intentionally hop out of their tank accidentally running into a booth. Maybe some form of temporary visual indicator (i.e. a 'beam' that shoots up occassionally, etc.) would be in order?

I could see the 'mini-missions' the commander gives a player being fun and it should add a strategic element to the game. These 'mini-missions' would also be a good way to increase a player's reputation. Seems like a win-win to me.

On to the 'nuke' idea. Will these be in the form of power ups (albeit extremely rare ones) or will they link with the gems system? If they're linked to the gem system and must be 'bought' by commanders, I could see these being used by commanders as 'game ending' weapons. It would certainly add some urgency to the game knowing the opposing commander is collecting gems in preparation for 'buying' a 'nuke'.


MadJack(Posted 2004) [#12]
GNS
For a map that blends tank action and FPS stuff, the FPS level geometry would be very simple and short. For more extensive FPS action, it would be dedicated single levels.

I think some sort of direction finder when on foot, pointing to the nearest booth would be required to reduce player frustration.

'accidentally running into a booth' - wouldn't happen very often and would be reduced by putting them in squares marked inacessible to tanks. And that the player might have to actively hit a button to activate them..

Nukes and linking them to gems. Yep, I had considered that with enough reputation, the player might be allowed more control in choosing gems -> units, and could have the option to 'buy' nukes (or lesser weapon upgrades). However, to do this well I think is beyond my time/energy. I really want to have this game either finished or well within the last quarter by the end of this year..

However, I could easily set up a scripted event whereby the enemy gets a nuke enabled tank and it's up to the player to intercept it.


GNS(Posted 2004) [#13]
All good ideas. With all of this TU talk I think I'll go play it...again. ;)


MadJack(Posted 2004) [#14]
Just spent some time over the weekend instituting a simple 'chat engine' - some simple code so that infantry units will strike up random conversations with each other if they're close enough, stationery, not in combat and not in a ranking unit's proximity..

Also set up some head turning code for infantry - so that infantry units turn their heads to look at each other and at the player as they pass by.

Amazing how much difference this makes to giving the impression that the player's presence is being noticed by AI units...


GNS(Posted 2004) [#15]
Very nice! Little things like those definitely make a big impact, atleast as far as I'm concerned.